We replaced our power supply (SMPS) by one based on EVM PMP6833 that uses UCC28600D regulator. Well, this time we were in the final test with first batch of equipment fitted with this new power supply (in this case the peristaltic infusion pump ST1000SET).
And we were surprised by a peculiar problem: the power supply turns itself off (and stays off for about 1-2 seconds) and then reconnects, but this occurs only with the equipment energized at 220VAC. We measure the voltage on the production line and it was within the tolerance source: 225V, stabilized.
In 110VAC there was no report on this behavior. But at 220VAC, we observed a rate of 30% equipment that exhibits this behavior. At that time, the production line is stopped, because we do not have confidence in release the equipment.
We are some days with equipment under analysis in our laboratory and we realize that the problem presents so well even peculiar.
Follow the obtained data:
The equipment fully operational consumes about 980 mA distributed to the stepper motor driver circuit (400 mA in the infusion flow rate applied in these tests), battery charging circuit (300 mA when the battery is fully discharged) and the general circuitry (display, microcontroller, etc.). We note that in this operating condition, the power supply works without anomaly.
However, when the current supplied decreases (stopping the infusion and consequently consumption), the power supply shows this behavior shutdown and restart. Analyzing the signals, we realize that the power supply, connected with a low current consumption (without motor + driver stepper motor and without load on battery), starts at the Green Mode. After a time (from what we saw in all power supply) it enters in the DCM mode. But we do not have absolute certainty about that and to us seems to be confusion in these operating modes, especially in the transition between them.
We already have tried here some things that had no effect (all based on the UCC28600D datasheet, in Design Considerations for the UCC28600 (SLUA399B) and UCC28600D Design Calculator (SLVC104) thereof:
- Turn on the MOSFET gate directly to pin 5 of UCC28600D;
- Change the capacitor of Soft Star (connected to pin 1 UCC28600D) to 100pF;
- Include a 10k resistor between the gate of the MOSFET and GND primary;
- Change the RPL from 1.5k to 1.8k;
- Connect the optocoupler collector directly to pin 2 of UCC28600D but keeping the capacitor;
- Withdraw the circuit TVS1 (this component was added because of our EMC test equipment needs).
When the power supply is energized (220VAC 60Hz) and after a few seconds it turns off and soon turns on. I believe they are calling the Green Mode because the FB pin voltage is 1.31V and after it turns off and turns on itself the voltage rises to 2.20 V, entering the DCM mode.
Something very curious is that these boards that exhibit this behavior if it is subjected to a greater load (in my case drained one 720mA of current using a resistor), the problem no longer appears, even at 220VAC.
I have some questions, besides of course need your help to better understand what is happening and mostly resolve this problem.
1 - What is the normal process of running this power supply? It is energized and enters Green Mode and then goes to another, or these modes can be fixed (the power supply working only in DCM, for example)?
2 - Always consider strange the time it takes to stabilize the output. From the moment power supply is energized, it takes about 2 to 3 seconds in 110VAC and 1 or 2 seconds at 220VAC for the voltage at the output to appear. This is the normal operation of it?
